Mirage auditing framework reveals that VFL unlearning methods passing output-level certification retain substantial class structure in representations, with no method achieving high utility plus both output and representation forgetting, plus class-sample asymmetry in residual traces.
Computers in biology and medicine132, 104319 (2021)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Do Vision Models Truly Forget? New Findings from Representation-Level Certification of Visual Unlearning in Vertical Federated Learning
Mirage auditing framework reveals that VFL unlearning methods passing output-level certification retain substantial class structure in representations, with no method achieving high utility plus both output and representation forgetting, plus class-sample asymmetry in residual traces.